Research on farmland water-saving irrigation system based on Internet of Things technology

1 Introduction
Water is the lifeline of agriculture, as well as the lifeline of the entire national economy and human life. The status of water resources and the level of water utilization have become important indicators for evaluating whether a country or a region can develop sustainably. China is a country with relatively scarce water resources. The average annual precipitation is 630 mm, which is lower than the precipitation of the global land surface and the Asian land surface; the average annual total freshwater resources are 2.8 trillion m3 ^, and the per capita water resources are only 2300 ^m3 , which is only 1/4 of the world's per capita level, ranking 109th in the world, and is one of the 13 countries with the poorest per capita water resources in the world; the water resources of cultivated land are 28500 m3/hm2 ^, which is 4/5 of the world average.

2 Current status of agricultural water use and development trend of water-saving irrigation
From the perspective of the total demand for water resources in the country, in the case of moderate drought, the total water demand in the country is about 550 billion m3 ^, and the water shortage is about 25 billion m3 ^. If unreasonable water supply factors such as groundwater over-exploitation and direct irrigation of sewage exceeding the standard are taken into account, the actual water shortage in the country is between 30 billion and 40 billion m3 ^. Agriculture is a major water user in China, accounting for about 73% of the total water consumption in the country, but the effectiveness is very poor, and the waste of water resources is very serious. The effective utilization rate of water in canal irrigation areas is only about 40%, and that in well irrigation areas is only about 60%. The grain production per m3 of ^water is less than 1 kg. In some developed countries, the effective utilization rate of water can reach more than 80%, and the grain production per m3 of ^water is generally more than 2 kg, among which Israel has reached 2.32 kg. This shows that the comprehensive application level of various water-saving agricultural technologies in China is still very low, and there is still a big gap compared with developed countries.
At present, the new water-saving irrigation technologies with great development potential are: First, crop regulation irrigation technology combined with biotechnology. Starting from the perspective of crop physiology, a certain degree of beneficial water deficit is actively applied in a certain period of time, so that crops experience beneficial water deficit training, improve quality, control the vigorous growth of the upper part, achieve dwarfing and dense planting, and achieve the purpose of water saving and increased production. Second, fine irrigation technology using 3S technology. It is to use the global satellite positioning system (GPS) and geographic information system (GIS), remote sensing technology (RS) and computer control system to obtain real-time information on the actual needs of crop growth in agricultural plots, and through information processing and analysis, the technology of watering crops on demand can maximize the utilization rate of water resources and the industrial rate of land. This is a hot spot in the development of farmland irrigation disciplines and an important part of the new agricultural technology revolution. Third, intelligent water-saving irrigation equipment technology. It is to integrate high-tech technologies such as biology, automatic control, microelectronics, artificial intelligence, and information science into water-saving irrigation machinery and equipment, detect soil and crop moisture in real time, and implement variable watering according to different water requirements of crops to achieve the best water-saving and increased production effect.
The farmland water-saving irrigation system based on Internet of Things technology designed in this paper organically combines the above three, and uses Internet of Things technology on this basis to realize a fully automated and information-based water-saving irrigation system.

3 System structure design
The farmland water-saving irrigation system consists of soil moisture sensors, IoT terminal acquisition units, sprinkler control terminals, and remote monitoring computer systems. As shown in Figure 1. The sensor is buried in the soil to directly obtain soil moisture information at various depths of 0 to 100 cm below the surface and convert it into 0 to 5V analog voltage signals. The IoT terminal acquisition unit is used to collect soil moisture information from the sensor on the one hand, and on the other hand, it uses the GPRS network mode to transmit the soil moisture information to the monitoring computer installed in the monitoring center. In a farmland water-saving irrigation monitoring system, there can be multiple IoT terminal acquisition units as needed. Each acquisition terminal can be used as a fixed soil moisture monitoring station, distributed at different feature points in the region to collect soil moisture information. The computer in the monitoring center cyclically receives the soil moisture information sent by each acquisition terminal. The monitoring computer analyzes and compares the received data with the crop water demand in the database to form the best irrigation plan. Then the monitoring computer sends the irrigation command to the sprinkler control terminal, and the sprinkler control terminal directly controls the sprinkler and deep well pumps and other equipment to perform irrigation operations. The system structure diagram is shown in Figure 1.

4 System Functions and Features
(1) System management: This part defines and maintains the structure of all data tables in the system; manages and maintains information such as accounts, permissions, interfaces, system operation parameters, file categories and attributes that maintain the normal operation of the system; and defines knowledge specifications in specific fields.
(2) Sprinkler control: Based on soil moisture information, the system formulates irrigation plans and remotely controls sprinklers through the GPRS network to achieve fully automatic irrigation.
(3) Data query and retrieval function: It has query and retrieval functions in various forms and ways, and outputs query results in the form of maps, tables or other forms. The query methods include point query, spatial query and logical condition combination query.
(4) Data acquisition unit automatic positioning: The terminal data acquisition automatically sends the latitude and longitude data to the monitoring center computer according to the placement location. The central computer automatically determines and displays the layout location of the data acquisition unit on the operation interface. (5) Data
analysis function: Different sections are analyzed for different attributes, and the results are provided in the form of thematic maps for printing.

5 Upper computer software structure
The monitoring center is mainly composed of a network server and a soil moisture data processing computer, which has a fixed Internet public network IP. Its function is to receive, process and display data in real time. The computer software of the monitoring center uses Asia Control KingView as the development platform. Through the secondary development of KingView, the central computer can collect and display data in real time, form databases and reports for irrigation forecasting and decision-making, calculate irrigation time and irrigation volume based on monitoring data, and display or print out monitoring and calculation results in charts, curves [1]. The system design will be based on the perspective of simplicity and ease of use. Its main operation interface is shown in Figure 2.

6 Design of IoT collection unit
The design of IoT collection unit is the terminal collection unit of this system. Due to the large detection range, large number, unfixed distribution and only used in the farming season, the collection terminal needs to be designed to be flexible and easy to install. Secondly, a GPS positioning module is installed on each collection terminal so that the data sent to the computer of the monitoring center is marked with a geographic location subscript. The central computer determines the specific geographic location of the collection point according to the geographic location subscript of the uploaded data, thereby realizing accurate data collection. In addition, since the data collection unit is placed in the farmland, the collection unit is powered by "solar panel + battery". [page] The collection
terminal is mainly composed of MCU unit, collection unit, solar power supply unit, communication unit, GPS positioning unit and other parts, and its structure is shown in Figure 3. Among them, the collection unit uses soil moisture temperature sensor to collect soil moisture data. After the data is processed by the embedded microcontroller MCU (MicroControl Unit), it is sent to the computer of the monitoring center through the GPRS network. The central computer collects temperature and humidity data and automatically displays relevant information. The signal output by the soil sensor is processed by the signal conditioning circuit and then transmitted to the analog-to-digital converter (ADC) inside the subsystem. The MCU starts the ADC at a fixed time, performs analog-to-digital conversion and takes away the data, then transmits the processed data to the GPRS module through the serial port, and starts the module to send the data to the GPRS wireless network. After the data is received by the GPRS network, it is forwarded to the Internet via the gateway and finally received by the central station computer connected to the Internet [2].

The core control MCU of the acquisition terminal is the core of the entire acquisition system. Considering the requirements of cost and processing performance, the embedded MCU uses the low-power 8-bit microprocessor ATmega128 produced by ATMEL as the processor chip of the data acquisition subsystem. The chip has rich hardware resources and has the advantages of low power consumption, multiple functions, low price and powerful performance. In this terminal, the core processor ATmega128 microcontroller is directly connected to the GPRS module through COM0 to complete the initialization of the GPRS module and the data transmission function based on the GPRS network. The GPS module in the system communicates through COM1 of ATmega128. ATmega128 itself has a 128K byte FLASH memory, and the lower computer program can be directly downloaded to the on-chip FLASH through the programmer. At the same time, ATmega128 has a 4K byte EEPROM memory, and the sensor acquisition data is directly stored in the EEPROM. The GPRS communication module and GPS module interfaces used in this design are both TTL level interfaces, which can be directly connected to the serial interface of the ATmega128 microcontroller. The interface circuit is shown in Figures 4 and 5:

The embedded GPRS module is powered by a DC 5V power supply. TXD and RXD are communication interfaces. In this design, they can be directly connected to the serial interface of the AVR microcontroller. ONLINE is an online indication interface. After connecting to the network, the port outputs a low-level signal, which drives the D1 light-emitting diode after reverse operation through 74ALS04. When the light-emitting diode lights up, it proves that the controller is now connected to the network. The GPS module is connected through the COM2 port of the microcontroller, as shown in Figure 5.
In this acquisition terminal, the soil moisture sensor interface is a 0-5V analog interface, so the sensor selects the JWSL-5VB protected temperature and humidity transmitter produced by Kunlun Coast Company. Its output signal is a DC voltage signal with a range of 0-5V. The temperature and humidity signals are output from their respective channels and are independent of each other. The signal output by the sensor is input to the ADC1 pin of ATmega128 after linear conversion processing, and the analog-to-digital conversion is performed by the ADC inside ATmega128. The ADC inside ATmega128 has 8 channels, each with a resolution of 10 bits and an input voltage range of 0 to 5V, which can meet the needs of the data collection of the system. The sensor signal conditioning and the interface circuit with ATmega128 are shown in Figure 6. After the voltage signal output by the sensor enters the circuit, it first passes through a low-pass filter. The voltage signal output by the sensor may have unstable factors. In addition, it is transmitted through a long cable and is also interfered by other devices during the process. A lot of medium and high frequency noise is superimposed on the signal. Therefore, before the signal enters the ADC of the processor, it is first filtered out as much as possible through a low-pass filter. Here, a first-order RC low-pass filter is used for filtering, with a cutoff frequency of 15.92Hz, which can effectively attenuate the medium and high frequency interference components and better reflect the changes in the signal. After the sensor output signal passes through the filter, it is buffered by a first-level voltage follower. After being converted into a 0-4.09V voltage signal by the voltage divider circuit composed of R1 and R3, it is buffered by a first-level buffer and finally sent to the ADC1 port of the processor (temperature signal is sent to ADC1, humidity signal is sent to ADC2) [2].

7 Conclusion
The water-saving irrigation control system based on the Internet of Things technology designed in this paper formulates the optimal irrigation plan according to the soil moisture and crop water demand, implements on-demand irrigation for crops, minimizes the consumption of water resources, alleviates the contradiction of increasingly tight water resources, and also provides a better growth environment for crops, fully develops the role of existing water-saving equipment, optimizes scheduling, and improves efficiency.

References:
[1] Chang Bo, Design of intelligent monitoring system for water-saving irrigation based on wireless sensor network [J]. Anhui Agricultural Science, 2010, 38(27): 23.
[2] Huang Weifeng et al., Embedded remote real-time monitoring system for forest soil temperature and humidity [J]. Agricultural Research, 2009, (12): 107-108.

Author profile: Zhao Hantao (1974-), male, senior engineer, engaged in scientific research in mechatronics and other fields.